Decision Making Overview eBook

41
Proprietary & confidential. © Decision Lens 2010 Decision Making For Better Results www.decisionlens.com 703-399-2100 February 2010 eBook

description

Decision making overview eBook provided by http://decisionlens.com/resources

Transcript of Decision Making Overview eBook

Page 1: Decision Making Overview eBook

Proprietary & confidential. © Decision Lens 2010

Decision Making For Better Results

www.decisionlens.com703-399-2100

February 2010

eBook

Page 2: Decision Making Overview eBook

Purpose

Successful decision making is one of the least understood or repeatable capabilities in most organizations. This eBook is about making better decisions through framing the problem, setting priorities, and establishing a defensible course of action. It is based on 30 years of advanced decision-making applications and experience, with many stories and examples

2

Page 3: Decision Making Overview eBook

Science and Decision Making

Why doesn’t our current scientific process work for decision making?

Science and technology use “analysis” Analysis breaks things down to provideinformation only, not necessarily the answersScience is empirical, while decision making seeks a solution from the synthesis of many partsScience is tangible, while decision making is more intangible

While science requires analysis, decision making needs more

3

Page 4: Decision Making Overview eBook

4

Example: we have successfully…

• Reached the shimmering moon.• Harnessed the energy of the atom.• Mastered global communication.• Invented the computer which spawned tens of

thousands of useful and not-so-useful things.

Page 5: Decision Making Overview eBook

5

But… Although man can travel to the moon, he is unable to devise a bettertransportation system on earth.

…Why?

Page 6: Decision Making Overview eBook

Transportation Example – Multi-facetedToday’s Problems Have Complexity

6

Not because the professionals involved are less imaginative or intelligent than those in the space program…

It is because transportation is more than just a technical problem. It is a “socio-technical”problem with many subtle facets:

There are many and diverse stakeholdersThere are many competing considerationsTransportation’s goals are not sharply defined.

Socio-technical problems are by and large provingmore subtle and difficult to handle.

??

?

?

????

?

?

??

Multi-facetedproblem

Page 7: Decision Making Overview eBook

Transportation Example – Competing Considerations: Most Decision Problems are “Multi-Criteria”

7

??

?

?

?

How do we accommodate the growing number of riders on our transportation system?

Can we continue to maintain and preservesuch a vast transportation network?

We need to reduce the transportation system’s impact on the environment!

Costs continue to spiral out of control. How do we reign them in?

How do we increase our partnerships andintegration with other transportation networks for themost efficiency and effectiveness?

Page 8: Decision Making Overview eBook

Transportation Example – Multiple Stakeholders: Most Decision Problems Have Competing Interests

8

PoliticiansRiders

Taxpayers

PlannersEmployees

Feds

Engineers

Environmentalists

Page 9: Decision Making Overview eBook

How can we arrive at decisions to solve this problem?

9

Politicians Engineers Riders

Taxpayers

Planners

Employees

Feds

Environmentalists

ComplexMulti-criteria questions

questions

questions

questions

stakeholders

Undefined goals

Social ramifications

Economics

Page 10: Decision Making Overview eBook

10

Page 11: Decision Making Overview eBook

Can’t we just use scientific analysis to figure it out?

Unfortunately, no. Complex Decisions Cannot Rely on Analysis Alone

11

Analysis is the backbone of science. It’s what got us to the moon.

But analysis alone cannot solve the problem of transportation.

Page 12: Decision Making Overview eBook

12

So how does one bring competing considerations together to form a decision?

Page 13: Decision Making Overview eBook

Decisions are based on judgments

Making judgments is one of the most basic of innate human skills.

By breaking up problems and then comparing discrete elements, we are able to arrive at a decision.

13

Even babies can discern between a smile and a frown. They are applying their innate ability to judge.

Page 14: Decision Making Overview eBook

Judgments about what?

14

Example: How do you choose a piece of fruit from a bowl of fruit?

Page 15: Decision Making Overview eBook

First, we identify characteristics

15

sweetness

tartness

texture

juiciness

size

Fruit characteristics:

Page 16: Decision Making Overview eBook

16

So how do we make judgments to compare the fruit?

We assign priority to those characteristics.

Can’t we just use numbers to assign these priorities, and be done with it?

No, because numbers themselves have no meaning.

Page 17: Decision Making Overview eBook

Aren’t numbers numbers? We have the habit to crunch numbers whatever they are…

17

An elderly couple looking for a town to which they might retire found Summerland, in Santa Barbara County, California, where a sign post read:

“Let’s settle here where there is a sense of humor,” said the wife; and they did.

CITY LIMIT

SummerlandPOPULATIONELEVATIONYEAR ESTABLISHED

30012081870

TOTAL: 5079

Page 18: Decision Making Overview eBook

18

So to select the fruit, we must prioritize the characteristics so they match our value system – our priorities

• Let’s compare two of the characteristics. Which are we hungrier for?

• The result is a measurable output of something that normally seems intangible – if Juicy is 3 times Sweet, we also know Sweet is 1/3 Juicy:

• Juicy: 3• Sweet: 1/3

Here we have decided that Juicy is 3 times more important than sweetness at the time (maybe we have a dry mouth)vs.

Page 19: Decision Making Overview eBook

Pairwise Comparisons of the Criteria are Used to Derive “Priorities”

• In decision making, we express judgments using a fundamental numerical scale (1 to 9).

• These represent our interpretation of dominance—one criterion may be twice as important (Moderate - 2) or five times (Strong - 5), all the way up to 9 times. A “1” is Equal importance.

• Comparing one factor to another gives its relative importance.

19

With respect to decision goal: To purchase a type of fruit, which is more important?

Juiciness Sweetness

Sweetness Tartness

Tartness Size

Size Filling

PRIORITY

Each criterion is compared to the others to assess the relative importance on a 1 to 9 scale

Page 20: Decision Making Overview eBook

“Relative Importance” enables us to Compare Tangible Criteria and Intangible Criteria

By definition, an intangible is something for which there is no scale of measurement.

Sweetness is an intangible factor that may differ significantly from one person to another.

Even “tangibles” such as “Size” use “made-up” measures“Size” is an invention. A unit of size such as an “inch” or a “pound” is an agreed upon measure of size. It has no meaning outside of our heads.“Cost” or “dollars” are similarly an agreed upon unit of value. They have no meaning outside of this agreed-upon value.

20

“Intangible” factors must be interpreted according to what they provide towards your goal. The power of pairwise comparisons is that it enables you

to trade-off both Tangible and Intangible factors against each other

Page 21: Decision Making Overview eBook

Comparing the Relative Importance of Each Criterion to All Others Gives their Weighted Values, called the “Priorities”

21

• The result of multiple sets of pairwise comparisons at each level is a weighted value hierarchy, with all of the priorities in the decision concisely captured and expressed as numerical values.

• These priorities will be the guidepost used to evaluate your alternatives (in this case fruit) later in the process.

Page 22: Decision Making Overview eBook

Now that we have our priorities we can select the fruit that best fits our need

22

Once you have established that juiciness is more important than sweetness by X amount, sweetness more than size, etcYou have to compare each alternative under the characteristic to establish which has more of that quality that you want…

Page 23: Decision Making Overview eBook

The decision:

23

APPLE!

Page 24: Decision Making Overview eBook

Now let’s go back to our more complex example of improving “transportation”

24

??

?

?

?

???

?

?

?

?

Multi-facetedproblem

Page 25: Decision Making Overview eBook

First, Because of the Complexity of the Criteria, We must Structure for Judgment

25

Decision making is about being clear, and articulating at various steps along the way. Once that is done, the relationshipsbetween elements is more easily formed and can therefore be structured for judgment.

1 We must be able to describe it.

2 We must be able to define the relationshipsbetween the parts.

3 We can then apply judgment to relate the parts according to a goal or purpose that we have in mind.

Page 26: Decision Making Overview eBook

How do we structure all of the competing questions for transportation?

26

Maintain/grow ridership

System preservation

Improve transportation and employee safety and service

Improve service integration

Improve patron safety

Improve the patron experienceExpand system capacity

Systems infrastructure rehab, renewal, and replacement

Facilities rehabilitation, renewal & replacement

Rolling stock and equipment rehabilitation, renewal, &

replacement

Improve accessibility (ridership)

Increase partnerships & economic development

Reduce costs and improve efficiency

We must create order from the chaos.

Page 27: Decision Making Overview eBook

Identify and find inter-related criteria

27

Maintain/grow ridership

System preservation

Improve transportation and employee safety and service

Improve service integration

Improve patron safety

Improve the patron experienceExpand system capacity

Facilities rehabilitation, renewal & replacement

Rolling stock and equipment rehabilitation, renewal, &

replacement

Improve accessibility (ridership)

Increase partnerships & economic development

Reduce costs and improve efficiency

Maintain/grow ridership

We must create order from the chaos.

Page 28: Decision Making Overview eBook

Identify and find inter-related criteria

28

Maintain/grow ridership

Maximize on-time performance

System preservation

Improve transportation and employee safety and service

Improve service integration

Improve patron safety

Improve the patron experienceExpand system capacity

Systems infrastructure rehab, renewal, and replacement

Facilities rehabilitation, renewal & replacement

Rolling stock and equipment rehabilitation, renewal, &

replacement

Improve accessibility (ridership)

Increase partnerships & economic development

Reduce costs and improve efficiency

Environmental enhancements

Page 29: Decision Making Overview eBook

“Synthesis” of Multiple Considerations

29

The considerations are framed as a hierarchy of inter-related criteria.

Maximize on-time performance

System preservationImprove transportation and employee safety and service

Improve service integration

Improve patron safety

Improve the patron experienceExpand system capacity

Systems infrastructure rehab, renewal, and replacementFacilities rehabilitation, renewal & replacementRolling stock and equipment rehabilitation, renewal, & replacement

Improve accessibility (ridership)

Increase partnerships & economic development

Reduce costs and improve efficiency

Environmental enhancements

In relative measurement a preference (judgment) is expressed on each pair of elements with respect to a common property they share – their “parent” element.

Maintain/grow ridership

Page 30: Decision Making Overview eBook

But first we must prioritize!

30

vs.

Patron experience On-time performance

• Let’s say that on-time performance is 5 times more important than patron experience with respect to Maintain/grow ridership.

• On-time performance: 5• Patron experience: 1/5

Page 31: Decision Making Overview eBook

Once we have pairwise compared all of the elements, we have our priorities

31

This is often called the “value model”, because we now have exact, quantified relative values for the entire picture to guide the decision.

Page 32: Decision Making Overview eBook

But if People Do a Poor Job of Comparing Elements, Don’t We Get “Garbage”?

The Analytic Hierarchy Process is very sophisticated. It tracks how consistenteach judgment pair is by looking at all other pairs.

If you said factor A was more important than factor B, and factor B more than factor C, that if you turn around and say “C more than A”, it knows that you have not been consistent.

Through a “matrix” of these comparisons, all of the pairs of judgments are evaluated at once.

It may be impossible to make a consistent set of judgments on some pieces that make them fit exactly with another consistent set of judgments on other related pieces. So we may neither be able to be perfectly consistent nor want to be.

We must allow for a modicum of inconsistency.

32

If inconsistency is over 10%, you may have poor judgments leading to a poor decision outcome.

Page 33: Decision Making Overview eBook

33

By using Priority, we’ve addressed how to make the judgments needed to

make decisions for the issue of Transportation.

But how to include multiple people with competing interests to form this

decision?

Compromise.

Page 34: Decision Making Overview eBook

Framing the Decision Problem

34

Not everyone would assign the same priority to the same need.• Some elements of the society will be unhappy no matter what is done. • A significant result of this divergence of goals is that the socio-technical

problems are not completely solvable in the same sense that going to the moon was.

What "solution" means in these cases is that a reasonable compromise among various requirements was achieved.

• The best solution may not be the best technical, or best economical, or best political, or best sociological solution, though it should have considered all these.

10 2 4 9 9 10

8.5

Page 35: Decision Making Overview eBook

Now, involve the Group:

Outcomes Are Dramatically Improved by Bringing Together Multiple Stakeholders for Their Judgments

35

PlannersEmployeesPoliticiansGov agenciesRidersEngineersTaxpayersEnvironmentalists

Robust decisions synthesize both multiple considerations and multiple viewpoints.

Page 36: Decision Making Overview eBook

Group Decision Making:

Synthesize the Judgments from All Participants as a Group Average (Geometric Mean)

36

Each criterion is compared to the others to assess the relative importance on a 1 to 9 scale

Group Average

Page 37: Decision Making Overview eBook

Similar to the fruit example, we use the resulting group priorities to evaluate projects…

37

Each project is evaluated for it’s contribution to the goals (criteria) you see here to the left. A project that does well on each criterion will generally have a higher value score than a project with limited (or no) contributions in each area.

Project 1 (Excellent)

Project 1 (Poor)

Project 1 (Fair)

Page 38: Decision Making Overview eBook

Rating the Projects Against The Criteria:

Aligning Your Investments with Your Strategy

38

The ratings scales are then used to rate each project on the value the project delivers for each criterion.

Rating scales are built for each criterion to rate initiatives on both quantitative and qualitative metrics. There is a score associated with each rating.

Page 39: Decision Making Overview eBook

Rating the Projects

39

Criteria priorities can then be changed as a “what-if” to see how that would impact the project priorities.

After the projects have been rated, they are displayed with their ratings scores, indicating alignment to the priorities.

“What if we madeSystem Preservation75% of the priority?”

Page 40: Decision Making Overview eBook

Why is the Analytic Hierarchy Process Easy to Use?

40

• It does not take for granted the measurements on scales, but asks that scale values be interpreted according to the objectives of the problem.

• It relies on elaborate hierarchic structures to represent decision problems and is able to handle problems of risk, conflict, and prediction.

• It can be used to make direct resource allocation, benefit/cost analysis, resolve conflicts, design and optimize systems.

It is an approach that describes how good decisions are made rather than prescribes how they should be made.

Page 41: Decision Making Overview eBook

Why the Analytic Hierarchy Process is Powerful in Corporate Planning

41

1) Forces one to face the entire problem at once.2) Breaks down criteria into manageable components.3) Leads a group into making a specific decision for consensus or tradeoff.4) Provides opportunity to examine disagreements and stimulate discussion and

opinion.5) Makes it possible to deal with conflicts in perception and in judgment.6) Interprets experience in a relevant way without reliance on a black box technique

like a utility function.7) Offers an actual measurement system. It enables one to estimate relative

magnitudes and derive ratio scale priorities accurately.8) Offers opportunity to change criteria, modify judgments.

It organizes, prioritizes and synthesizes complexity within a rational framework.